In biotechnology, separation and analysis of chemical samples is critically important. Moreover, it is desirable to conduct multiple separations and analyses of the separated components simultaneously to increase the speed and efficiency at which chemical samples are evaluated. For example, separation technologies such as electrophoresis are used in DNA sequencing, protein molecular weight determination, genetic mapping, and other types of processes used to gather large amounts of analytical information about particular chemical samples.
One method used to separate chemical samples into their component parts is electrophoresis. Electrophoresis is the migration of charged colloidal particles or molecules through a solution under the influence of an applied electric field usually provided by immersed electrodes, where the colloidal particles are a suspension of finely divided particles in a continuous medium.
Historically, a polymer gel containing the finely divided particles is placed between two glass plates and an electric field applied to both ends of the plates. This method, however, offers a low level of automation and long analysis times.
More recently, the capillary electrophoresis (hereafter “CE”) method was developed, which has the added advantages of speed, versatility and low running costs. Operation of a CE system involves application of a high voltage (typically 10-30 kV) across a narrow bore capillary (typically 25-100 Fm). The capillary is filled with electrolyte solution which conducts current through the inside of the capillary. The ends of the capillary are dipped into reservoirs filled with the electrolyte. Electrodes made of an inert material such as platinum are also inserted into the electrolyte reservoirs to complete the electrical circuit. A small volume of sample is injected into one end of the capillary. The capillary passes through a detector, usually a UV absorbance detector, at the opposite end of the capillary. Application of a voltage causes movement of sample ions towards their appropriate electrode usually passing through the detector. Different sample ions arrive at a detection part of the capillary at different times. The sample may be labeled with a fluorescent marker so that when the sample passes through a beam of light at the detector, the fluorescent marker fluoresces and the fluorescence is detected as an electric signal. The intensity of the electric signal depends on the amount of fluorescent marker present in the detection zone. The plot of detector response with time is then generated which is termed an electropherogram.
CE is a particularly preferred separation method, as it allows the use of high electric fields, due to the capillary tube efficiently dissipating the resulting heat produced by the electric field. As such, the separations achieved are much better than the more traditional electrophoretic systems. In addition, multiple capillary tubes may be closely spaced together and used simultaneously to increase sample throughput.
In traditional CE systems, analysis or detection of the separated components is performed while the sample is still located within the capillary, and may be accomplished using photometric techniques such as adsorbance and fluorescence. Adsorbance and fluorescence is where excitation light is directed toward the capillary tube, and light emitted from the sample (e.g., fluorescence) is measured by a detector, thereby providing information about the separated components. Therefore, in these systems, excitation light directed at the sample, as well as light emitted from the sample, must be transmitted through the capillary's walls. A drawback of this approach is that the fused silica capillaries typically used in capillary electrophoresis are poor optical elements and cause significant scattering of light. The problem associated with light scattering is exacerbated by having multiple capillaries disposed side-by-side, as scattered excitation light from one capillary interferes with the detection of samples in neighboring capillaries.
One approach to solving the problem of on-capillary detection has been to detect a sample after the sample emerges from the capillary in a detection cell having superior optical characteristics, e.g., a flat quartz chamber. In this system, a sample is transported from the outlet of a capillary to the detection cell by electrophoresis under the influence of the same voltage difference used to conduct the electrophoretic separation. Examples of this type of system are disclosed in U.S. Pat. No. 5,529,679, which is incorporated herein by reference.
A variation of the above system replaces the capillary tubes with a series of parallel channels formed in a plate or chip, where the channels are in fluid communication with a detection cell in a manner similar to that described above. This type of system is known as a micro-channel array.
While addressing some of the abovementioned problems, the detection cell type CE system has drawbacks of its own. For example, excitation energy, such as light from a laser, has the tendency to scatter, thereby diminishing the energy's intensity as it transmitted through the detection cell.
A partial cross-section of a prior art detection cell 102 is shown in FIG. 1A. The detection cell 102, typically made from glass, forms a cavity 108, which is filled with an electrolytic polymer 110 containing a sample to be detected. Rays of light 104, typically from a laser, enter the detection cell 102 at a first end 112. Because the first end 112 is normal to the rays of light 104, the light 104 does not scatter, i.e., reflect or refract, when passing into the detection cell, from air to glass. However, when the light 104 passes through the boundary 106 between the detection cell and the polymer 110, the light is refracted. This is due to the angle or slope of the boundary 106, and the difference in refractive indices of the glass and polymer. The angle or slope of the boundary 106 is caused by current etching and mastering technologies, which are unable to produce optically flat vertical cavity walls in glass or plastic.
The refracted light obeys the law of refraction, i.e.,RI1 sin(A1)=RIR sin(AR)Where                RI1=first refractive index;        A1=angle of incidence;        RIR=second refractive index; and        AR=angle of refraction.        
As the polymer has a refractive index (approximately 1.41) less than the refractive index of glass (approximately 1.52), the angle of refraction is larger than the angle of incidence and the light bends further away from the normal to the boundary 106. Many light rays are lost due to light escaping 116 out of the detection cell instead of being trapped in the cavity by Fresnel reflection. This degrades the intensity of excitation light incident on the samples, which in turn adversely effects the detected signal strength. Furthermore, refracted light rays may also reflect 114 off the internal surfaces of the cavity 108 causing interference and detection signal loss. In other words, the curved or angled interfaces or boundaries in combination with the unfavorable refractive index change at the glass to polymer boundary or interface, leads to unsatisfactory light intensity and quality, leading to poor sample detection.
Moreover, the optical channels of these systems are often extended along a narrow tunnel to a dead end at the light receiving side of the detection cell. This introduces an interface, at the point where the tunnel meets the detection cell, that is hard to control, keep clean, and isolate from the high voltages in the detection cell. Further, the interface may cause a distortion of the electrophoresis field. Still further, bubbles trapped in the tunnel may optically interfere with the excitation light. This, in combination with poor light intensity and quality, further aggravates the signal detection.
In light of the above drawbacks, there is a need for an improved detection cell that provides better control over excitation light that is used to detect or analyze separated sample components that have been produced using techniques such as CE tube or microchip technology. Further, there is a need for an improved method for controlling the direction of the light rays within the detection cell.